Releasing Persimmon-8B

Open paper
Benchmark Model Rank Results
common-sense-reasoning-on-arc-easyPersimmon-8B-FT#76Accuracy: 64.0
code-generation-on-humanevalPersimmon-8B-FT#136Pass@1: 20.7
common-sense-reasoning-on-winograndePersimmon-8B-FT#153Accuracy: 54.6

Rank counts only results with a code link.